The Cancer Genome Atlas (TCGA) project holds promise for a comprehensive understanding of human cancer through the application of genomic technologies. However, current cancer genomic analytical and visualization technologies still have many limitations that will likely prevent investigators from taking full advantage of this resource. The proposed UCSC-Buck Institute Genome Data Analysis Center will support an integrative analysis of TCGA data for all surveyed cancer types throughout the project. The major components of the pipeline are a pathway-centric multi-layer machine learning tool called Biolntegrator, a genome rearrangement detector for next-gen sequencing data, and the tightly coupled UCSC browser tool suite. We aim to detect cancer-associated molecular alterations and the biological pathways that are perturbed by them in tumor samples. Samples will then be classified into clinically relevant categories based on pathway perturbations rather than perturbations of individual genes, which we believe will be more robust, biologically meaningful and clinically accurate. Using Biolntegrator and the associated tools, we will further integrate TCGA data with datasets from external studies, including cell line studies, animal studies and clinical trials, to identify (1) cancer-associated molecular alterations;(2) dysregulated pathways and signatures useful in clinical diagnosis, prognosis, and drug response prediction;and (3) gene targets for the development of novel therapeutics. These results will provide the basis for a refined patient stratification in therapy and will generate new hypotheses for translational research. The tightly coupled UCSC browser suite, which will be enhanced to accommodate the needs of the TCGA project, includes the UCSC Cancer Genomics Browser for visualizing TCGA cancer genomics, clinical data, and analysis results;the UCSC Tumor Browser for displaying tumor genome rearrangements and other tumor mutations;and the UCSC Human Genome Browser for integrating the data with human genome annotations and information gleaned from other projects such as ENCODE and the NIH Epigenomics Roadmap Initiative. The browser resource, hosting this rapidly growing body of cancer genomics data, will enable investigators to perform interactive in-silico experiments to test new hypotheses derived from the TCGA data. Collectively, these proposed tools will enable cancer researchers to better explore the breadth and depth of the TCGA resources and to further characterize molecular pathways that influence cellular dynamics and stability in cancer. Ultimately, insights gained by applying these tools will advance our knowledge of human cancer biology and stimulate the discovery of new prognostic and diagnostic markers, leading to new therapeutic and preventative strategies.